TY - GEN
T1 - Noise-assisted Nonlinear Squeezing Transform for Rotating Machinery Fault Diagnosis
AU - Zheng, Yiming
AU - Wang, Shibin
AU - Ding, Baoqing
AU - Lin, Shujing
AU - Chen, Xuefeng
AU - Gao, Robert X.
AU - Yan, Ruqiang
N1 - Publisher Copyright:
© 2025 IEEE.
PY - 2025
Y1 - 2025
N2 - This paper proposes a noise-assisted nonlinear squeezing transform (NANST) technique based on the spectrogram zeros (SPZ). This method can effectively weaken the relative amplitude of noise and thus enhance signal characteristics for rotating machinery fault diagnosis. The statistical characteristics of the zeros are analyzed, and it is found that the zeros of noise have random distribution characteristics, and the signal of interest has fixed structure and fixed zeros. Thus, the results of multiple nonlinear squeezing transform (NST) results with different noises are fused by the geometric mean to average and weaken the noise effect. Therefore, the results with clearer time-frequency representation (TFR) can be obtained, even for weak signals. The performance of the proposed method is verified in the rotating machinery high-speed bearing experiment.
AB - This paper proposes a noise-assisted nonlinear squeezing transform (NANST) technique based on the spectrogram zeros (SPZ). This method can effectively weaken the relative amplitude of noise and thus enhance signal characteristics for rotating machinery fault diagnosis. The statistical characteristics of the zeros are analyzed, and it is found that the zeros of noise have random distribution characteristics, and the signal of interest has fixed structure and fixed zeros. Thus, the results of multiple nonlinear squeezing transform (NST) results with different noises are fused by the geometric mean to average and weaken the noise effect. Therefore, the results with clearer time-frequency representation (TFR) can be obtained, even for weak signals. The performance of the proposed method is verified in the rotating machinery high-speed bearing experiment.
KW - Fault diagnosis
KW - Time-frequency analysis
KW - Zeros of the spectrogram
UR - https://www.scopus.com/pages/publications/105012194752
U2 - 10.1109/I2MTC62753.2025.11079122
DO - 10.1109/I2MTC62753.2025.11079122
M3 - 会议稿件
AN - SCOPUS:105012194752
T3 - Conference Record - IEEE Instrumentation and Measurement Technology Conference
BT - IEEE International Instrumentation and Measurement Technology Conference, I2MTC 2025 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2025 IEEE International Instrumentation and Measurement Technology Conference, I2MTC 2025
Y2 - 19 May 2025 through 22 May 2025
ER -